Report Date: 2026-03-16
Analysis Window: 30 days
Languages Tracked: python, typescript, rust, go, javascript
Top N: 50
Overview
根据过去30天的数据分析，GitHub AI领域的开源项目趋势明确指向了“Agent”这一核心主题。Agent类项目在数量、综合评分和持续热度上均占据主导地位，显示出该方向已成为当前AI技术落地和创新的主要焦点。在技术栈方面，TypeScript和Python是主流开发语言，同时Rust和Go等高性能语言在基础设施项目中也扮演着重要角色。
Top 50 Repositories
| Rank | Repository | Category | Language | Heat_7 | Heat_30 | Score |
|---|---|---|---|---|---|---|
| 1 | openclaw | typescript | 0 | 121,912 | 406 | |
| 2 | pi-mono | agent | typescript | 0 | 11,873 | 302 |
| 3 | shannon | agent | typescript | 0 | 11,531 | 38 |
| 4 | learn-claude-code | agent | typescript | 4,067 | 10,552 | 732 |
| 5 | OpenViking | agent | python | 6,514 | 10,104 | 1,434 |
| 6 | AstrBot | llm | python | 5,041 | 8,989 | 246 |
| 7 | agent-browser | agent | rust | 2,349 | 8,198 | 327 |
| 8 | OpenSandbox | agent | python | 0 | 6,794 | 22 |
| 9 | BitNet | llm | python | 6,028 | 6,368 | 279 |
| 10 | marketingskills | agent | javascript | 1,746 | 5,842 | 94 |
| 11 | promptfoo | llm | typescript | 5,337 | 5,808 | 248 |
| 12 | superset | agent | typescript | 0 | 5,429 | 18 |
| 13 | codex | agent | rust | 1,663 | 5,304 | 88 |
| 14 | skills | agent | typescript | 0 | 4,377 | 14 |
| 15 | notebooklm-py | agent | python | 2,266 | 3,864 | 109 |
| 16 | ai-hedge-fund | python | 2,322 | 3,375 | 110 | |
| 17 | litellm | llm | python | 0 | 3,190 | 10 |
| 18 | RuVector | rust | 0 | 2,955 | 9 | |
| 19 | rowboat | typescript | 0 | 2,808 | 9 | |
| 20 | anything-llm | javascript | 0 | 1,794 | 5 | |
| 21 | gh-aw | agent | go | 0 | 1,610 | 5 |
| 22 | agents | agent | typescript | 0 | 1,337 | 4 |
| 23 | SillyTavern | llm | javascript | 0 | 1,335 | 4 |
| 24 | worktrunk | agent | rust | 405 | 1,270 | 21 |
| 25 | oh-my-pi | agent | typescript | 0 | 1,215 | 4 |
| 26 | cli | agent | go | 0 | 1,193 | 3 |
| 27 | one-api | llm | javascript | 364 | 968 | 18 |
| 28 | qdrant | rust | 0 | 927 | 3 | |
| 29 | agent-deck | llm | go | 0 | 731 | 2 |
| 30 | figma-console-mcp | typescript | 0 | 678 | 2 | |
| 31 | ai-dev-kit | agent | python | 0 | 628 | 2 |
| 32 | rig | llm | rust | 189 | 574 | 10 |
| 33 | mcp_excalidraw | llm | javascript | 0 | 539 | 1 |
| 34 | docker-agent | agent | go | 394 | 487 | 18 |
| 35 | skills | agent | javascript | 0 | 447 | 1 |
| 36 | firecrawl-mcp-server | llm | javascript | 0 | 288 | 0 |
| 37 | atlassian-mcp-server | agent | javascript | 0 | 119 | 0 |
| 38 | impeccable | javascript | 7,077 | 0 | 684 | |
| 39 | page-agent | agent | typescript | 6,971 | 0 | 631 |
| 40 | InsForge | agent | typescript | 0 | 0 | 309 |
| 41 | GitNexus | agent | typescript | 0 | 0 | 270 |
| 42 | deepagents | agent | python | 0 | 0 | 266 |
| 43 | hermes-agent | agent | python | 5,152 | 0 | 220 |
| 44 | dimos | agent | python | 0 | 0 | 193 |
| 45 | cognee | agent | python | 0 | 0 | 162 |
| 46 | hindsight | agent | python | 1,564 | 0 | 123 |
| 47 | project-nomad | typescript | 0 | 0 | 123 | |
| 48 | chrome-devtools-mcp | agent | typescript | 0 | 0 | 111 |
| 49 | Fabric | go | 0 | 0 | 90 | |
| 50 | plannotator | agent | typescript | 0 | 0 | 42 |
Category Breakdown
(10 projects)
未分类项目（10个）涵盖了AI基础设施（如向量数据库qdrant）和创新的AI应用（如ai-hedge-fund），它们作为AI生态系统的支撑和延伸，虽然热度稍逊，但同样是不可或缺的组成部分。
Average Heat_7: 939
Average Score: 144
- openclaw - Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
- ai-hedge-fund - An AI Hedge Fund Team
- RuVector - RuVector is a High Performance, Real-Time, Self-Learning, Vector Graph Neural Network, and Database built in Rust.
- rowboat - Open-source AI coworker, with memory
- anything-llm - The all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configration.
- qdrant - Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
- figma-console-mcp - Your design system as an API. Connect AI to Figma for extraction, creation, and debugging.
- impeccable - The design language that makes your AI harness better at design.
- project-nomad - Project N.O.M.A.D, is a self-contained, offline survival computer packed with critical tools, knowledge, and AI to keep you informed and empowered—anytime, anywhere.
- Fabric - Fabric is an open-source framework for augmenting humans using AI. It provides a modular system for solving specific problems using a crowdsourced set of AI prompts that can be used anywhere.
agent (30 projects)
Agent是本次趋势分析中绝对的核心，占据了Top 50榜单中的30个席位，并拥有最高的平均分（185.4分），表明该领域的项目不仅数量众多，而且影响力巨大，代表了AI领域最活跃的开发前沿。
Average Heat_7: 1,103
Average Score: 185
- pi-mono - AI agent toolkit: coding agent CLI, unified LLM API, TUI & web UI libraries, Slack bot, vLLM pods
- shannon - Shannon Lite is an autonomous, white-box AI pentester for web applications and APIs. It analyzes your source code, identifies attack vectors, and executes real exploits to prove vulnerabilities bef...
- learn-claude-code - Bash is all you need - A nano Claude Code–like agent, built from 0 to 1
- OpenViking - OpenViking is an open-source context database designed specifically for AI Agents(such as openclaw). OpenViking unifies the management of context (memory, resources, and skills) that Agents need th...
- agent-browser - Browser automation CLI for AI agents
- OpenSandbox - OpenSandbox is a general-purpose sandbox platform for AI applications, offering multi-language SDKs, unified sandbox APIs, and Docker/Kubernetes runtimes for scenarios like Coding Agents, GUI Agent...
- marketingskills - Marketing skills for Claude Code and AI agents. CRO, copywriting, SEO, analytics, and growth engineering.
- superset - IDE for the AI Agents Era - Run an army of Claude Code, Codex, etc. on your machine
- codex - Lightweight coding agent that runs in your terminal
- skills - The open agent skills tool - npx skills
- notebooklm-py - Unofficial Python API and agentic skill for Google NotebookLM. Full programmatic access to NotebookLM's features—including capabilities the web UI doesn't expose—via Python, CLI, and AI agents ...
- gh-aw - GitHub Agentic Workflows
- agents - Build and deploy AI Agents on Cloudflare
- worktrunk - Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows
- oh-my-pi - ⌥ AI Coding agent for the terminal — hash-anchored edits, optimized tool harness, LSP, Python, browser, subagents, and more
- cli - Entire is a new developer platform that hooks into your git workflow to capture AI agent sessions on every push, unifying your code with its context and reasoning.
- ai-dev-kit - Databricks Toolkit for Coding Agents provided by Field Engineering
- docker-agent - AI Agent Builder and Runtime by Docker Engineering
- skills - A collection of AI agent skills for working with Expo projects and Expo Application Services
- atlassian-mcp-server - Remote MCP Server that securely connects Jira and Confluence with your LLM, IDE, or agent platform of choice.
- page-agent - JavaScript in-page GUI agent. Control web interfaces with natural language.
- InsForge - Give agents everything they need to ship fullstack apps. The backend built for agentic development.
- GitNexus - GitNexus: The Zero-Server Code Intelligence Engine - GitNexus is a client-side knowledge graph creator that runs entirely in your browser. Drop in a GitHub repo or ZIP file, and get an interactive ...
- deepagents - Agent harness built with LangChain and LangGraph. Equipped with a planning tool, a filesystem backend, and the ability to spawn subagents - well-equipped to handle complex agentic tasks.
- hermes-agent - The agent that grows with you
- dimos - Dimensional is the agentic operating system for physical space. Vibecode humanoids, quadrupeds, drones, and other hardware platforms in natural language and build multi-agent systems that work seam...
- cognee - Knowledge Engine for AI Agent Memory in 6 lines of code
- hindsight - Hindsight: Agent Memory That Learns
- chrome-devtools-mcp - Chrome DevTools for coding agents
- plannotator - Annotate and review coding agent plans and code diffs visually, share with your team, send feedback to agents with one click.
llm (10 projects)
LLM相关项目虽然数量不多（10个），但拥有最高的平均近期热度（1695.9），这主要得益于像microsoft/BitNet这样的模型底层创新项目。这表明围绕大语言模型本身的基础研究和优化工具依然是社区关注的热点。
Average Heat_7: 1,695
Average Score: 81
- AstrBot - Agentic IM Chatbot infrastructure that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨
- BitNet - Official inference framework for 1-bit LLMs
- promptfoo - Test your prompts, agents, and RAGs. Red teaming/pentesting/vulnerability scanning for AI. Compare performance of GPT, Claude, Gemini, Llama, and more. Simple declarative configs with command line ...
- litellm - Python SDK, Proxy Server (AI Gateway) to call 100+ LLM APIs in OpenAI (or native) format, with cost tracking, guardrails, loadbalancing and logging. [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthr...
- SillyTavern - LLM Frontend for Power Users.
- one-api - LLM API 管理 & 分发系统，支持 OpenAI、Azure、Anthropic Claude、Google Gemini、DeepSeek、字节豆包、ChatGLM、文心一言、讯飞星火、通义千问、360 智脑、腾讯混元...
- agent-deck - Terminal session manager for AI coding agents. One TUI for Claude, Gemini, OpenCode, Codex, and more.
- rig - ⚙️🦀 Build modular and scalable LLM Applications in Rust
- mcp_excalidraw - MCP server and Claude Code skill for Excalidraw — programmatic canvas toolkit to create, edit, and export diagrams via AI agents with real-time canvas sync.
- firecrawl-mcp-server - 🔥 Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients.
Dark Horse Projects
本周期涌现出多匹高分“黑马”，凸显了AI领域的快速创新。其中，volcengine/OpenViking（Agent上下文数据库）、alibaba/page-agent（页面GUI Agent）和pbakaus/impeccable（AI设计语言）表现尤为突出。这些项目分别代表了Agent专用基础设施、人机交互自动化以及AI与设计结合等新兴方向，它们的迅速崛起预示着这些细分赛道可能成为下一阶段的增长点。
| Repository | Score | Heat_30 | Heat_7 | Category |
|---|---|---|---|---|
| openclaw | 406 | 121,912 | 0 | |
| pi-mono | 302 | 11,873 | 0 | agent |
| learn-claude-code | 732 | 10,552 | 4,067 | agent |
| OpenViking | 1,434 | 10,104 | 6,514 | agent |
| AstrBot | 246 | 8,989 | 5,041 | llm |
| agent-browser | 327 | 8,198 | 2,349 | agent |
| BitNet | 279 | 6,368 | 6,028 | llm |
| promptfoo | 248 | 5,808 | 5,337 | llm |
| notebooklm-py | 109 | 3,864 | 2,266 | agent |
| ai-hedge-fund | 110 | 3,375 | 2,322 | |
| impeccable | 684 | 0 | 7,077 | |
| page-agent | 631 | 0 | 6,971 | agent |
| InsForge | 309 | 0 | 0 | agent |
| GitNexus | 270 | 0 | 0 | agent |
| deepagents | 266 | 0 | 0 | agent |
| hermes-agent | 220 | 0 | 5,152 | agent |
| dimos | 193 | 0 | 0 | agent |
| cognee | 162 | 0 | 0 | agent |
| hindsight | 123 | 0 | 1,564 | agent |
| project-nomad | 123 | 0 | 0 | |
| chrome-devtools-mcp | 111 | 0 | 0 | agent |
Consecutive Appearances
榜单上的“常客”项目进一步印证了Agent方向的长期性和稳定性。openclaw/openclaw、badlogic/pi-mono等项目已连续上榜8周之久，说明它们已成为开发者社区中的基础工具或重要参考。这些项目的持续热门表明，AI Agent并非短暂热潮，而是具有深远影响的技术范式，其生态正在逐步成熟和巩固。
| Repository | Weeks in Top | Category | Current Heat_7 |
|---|---|---|---|
| openclaw | 8 | 0 | |
| pi-mono | 8 | agent | 0 |
| shannon | 8 | agent | 0 |
| learn-claude-code | 2 | agent | 4,067 |
| OpenViking | 2 | agent | 6,514 |
| AstrBot | 4 | llm | 5,041 |
| OpenSandbox | 3 | agent | 0 |
| superset | 4 | agent | 0 |
| codex | 8 | agent | 1,663 |
| ai-hedge-fund | 2 | 2,322 | |
| litellm | 4 | llm | 0 |
| anything-llm | 3 | 0 | |
| gh-aw | 2 | agent | 0 |
| agents | 4 | agent | 0 |
| SillyTavern | 2 | llm | 0 |
| worktrunk | 2 | agent | 405 |
| qdrant | 2 | 0 | |
| agent-deck | 4 | llm | 0 |
| ai-dev-kit | 2 | agent | 0 |
| rig | 4 | llm | 189 |
| skills | 4 | agent | 0 |
| impeccable | 2 | 7,077 | |
| page-agent | 2 | agent | 6,971 |
| GitNexus | 4 | agent | 0 |
| hermes-agent | 2 | agent | 5,152 |
| chrome-devtools-mcp | 5 | agent | 0 |
Highlighted Repositories
openclaw/openclaw
该项目以极高的30天热度（121912）和连续8周上榜的记录稳居榜首，已成为AI Agent领域的标杆。作为一个跨平台的个人AI助理，它不仅自身热度高，还带动了生态发展，榜单上多个项目都围绕其进行构建或将其作为对标产品。尽管其最近7天热度为0，但这反映了其主要增长爆发期位于30天窗口的前段，其积累的巨大影响力使其地位依然稳固。
volcengine/OpenViking
作为专为AI Agent设计的上下文数据库，OpenViking凭借其高达1434的评分和6514的近期热度，精准地切入了当前Agent开发中的核心痛点——状态和记忆管理。它通过文件系统的范式来统一管理上下文，为Agent实现长期记忆和技能演进提供了创新的解决方案，显示出Agent基础设施正在走向专业化和底层化。
microsoft/BitNet
微软官方发布的1-bit大语言模型推理框架，近期热度（6028）极高，反映了业界对极致模型压缩和效率提升的强烈兴趣。该项目旨在从根本上降低LLM的算力和内存需求，是推动大模型在边缘设备和低资源环境中普及的关键技术探索，代表了LLM领域一个重要的前沿研究方向。
alibaba/page-agent
阿里巴巴推出的page-agent是一个近期热度（6971）和评分（631）双高的黑马项目，专注于通过自然语言直接操控网页GUI。它解决了AI自动化与现有Web生态系统交互的“最后一公里”问题，是Agent技术在前端自动化和人机交互领域极具潜力的应用，预示着未来网页浏览和操作方式的变革。
promptfoo/promptfoo
此项目是一个面向Prompt、Agent和RAG系统的测试评估工具，其高关注度表明AI开发正在进入更加成熟的工程化阶段。随着AI系统日趋复杂，对其进行系统性测试、红队评估和漏洞扫描的需求日益迫切，promptfoo的 流行恰好满足了LLMOps领域对应用可靠性和安全性的核心诉求。
Methodology
Data Sources:
- GitHub Trending pages
- GitHub API stargazer timestamps
Metrics:
- Heat_7: Stars gained in last 7 days
- Heat_30: Stars gained in last 30 days
- Score: Weighted scoring combining short-term heat and sustained growth
- Formula: 0.6 × stars_1d + 0.3 × (stars_7d / 7) + 0.1 × (stars_30d / 30)
- Emphasizes recent activity (60%) while considering sustained trends (40%)
Filtering:
- Include keywords: %!s(int=30)
- Exclude keywords: %!s(int=30)
- Categories: %!s(int=30)
Ranking:
- Sort by Heat_30 (descending)
- Tie-break by Score (descending)
- Select top %!d(string=ai, llm, agent, gpt, rag, langchain, machine-learning, deep-learning, neural, transformer, embedding, vector)
Limitations:
- Trending data limited to GitHub's trending algorithm
- Star history may be incomplete for repos with >40k stars (API pagination limits)
- LLM-generated commentary is interpretive, not prescriptive
- Weekly snapshots may miss short-lived trends%!(EXTRA string=tutorial, awesome, learning, course, book, string=vision, infra, agent, llm, rag, int=50)
Generated at: 2026-03-16T00:32:40Z